Motorin'

I did some research online, trying to figure out how I'm going to interface the motors with the Arduino Nano. After my previous expeditions, I came to the conclusion that I would have to fiddle around with pulse-width-modulation (PWM) in order to properly control the motors. Fortunately, the Nano has 6 outputs that are capable of emulating PWM, and about 8 analog inbputs that could (possibly) be used to read PWM signals from the receiver I'll eventually get. Some sources mentioned that I would need to put an H-bridge between the motors and the microcontroller, though, and I'm wondering if a speed controller would work just as fine.

I got a little antsy, and decided to try flying the Draganflyer again. The wind from earlier today had settled down a little, but it was still incredibly cold. Despite this, Squishy and Jason joined me for the flight test. After a few warm-up flights, I went ahead and had Squishy try filming it with my cellphone like last Friday, and this time it didn't crash. Since I was flying with the attitude stabilization system on, it was easy-peasy even when breazy. Though it's difficult to see in the video, the wind actually twisted one of the motors at an angle, and it still flew level! I think that the attitude system was adjusting for any changes the twisted rotor was causing, so the rotors all sped-up and sped-down to compensate. This didn't provide any translational stability, of course, and so the Draganflyer often ended up veering off considerably to the left or right. We played around with it a little bit... needless to say it ate dirt a couple times. Luckily, the blades are made out of a flexible plastic, so even though they bend easily, they return to form just as readily. I do have to clean some dirt out of the gears, though.

I'm definitely going ahead with the LED idea, but I'm afraid of taking some of the LED's from the xUFO. They're soldered onto circuit boards underneath the motors, and I'm not sure how integral they are to the circuit (they're the kind that are actually built-in to the PCB). I might just grab some LED's I have at home and use those -- anything to help distinguish the front end of the Draganflyer from the other parts. It's time we did something about symmetry...

TL;DR

• ESC or H-bridge: can an esc be used instead of trying to setup a circuit with an H-bridge for motor control
• Draganflyer is one tough mother-: shut yo' mouth! The damn thing can fly level when its motors are off-kilter, and its blades can bend back into shape; I'm still getting the replacement kit, though
  
• Rotor RPM testing: figure out the changes in rotor RPM with certain inputs, like roll right, roll left, pitch up, etc
  

Litany of Durability

I decided to take out the Draganflyer for a spin again this past Friday, this time with a couple witnesses to take video. I wanted video of the rotorcraft with artificial stability off and on for comparison. Of course, irony being the way it is, I managed to crash the rotorcraft spectacularly while flying with artificial stability off. I say spectacularly because parts flew off, making it a big show. In actuality, only one of the rotor blades flew off when it struck the ground. I always thought that the nylon screws used to hold it in were to prevent damage to the plastic gears they bore into -- it seems they serve a dual-purpose, as the screw-heads popped off when the blade impacted the ground, ejecting the blade and preventing any further damage to it or the motor. I bid my friends adieu (they had to leave) and ran back to the lab, praying that there were spare screws available. Fortunately, there were, and after a little tweaking and brushing off some dirt, the Draganflyer looked almost like new.

I ran outside to fly it one more time that day, without anyone taking video. I flipped the stability switch on this time, and sure enough, it flew well. It's not like I didn't expect it -- I just wanted to verify that it was difficult to fly without the Draganflyer thermal stability system first, and my expectations were vindicated by the crash earlier that day. It didn't go wildly out of control, though; in fact, it could have easily been chalked up to pilot error until I flew it with stability on. Basically, flying without attitude control, the Draganflyer began to roll right, causing it to lose altitude and slap itself into the ground. With attitude control on, it still slips and slides around in the air, but it doesn't roll or pitch as greatly or quickly with attitude control active. In my haste I didn't grab anyone to take video this time, so I filmed a couple hops by holding my cellphone to the RC transmitter. It wasn't long before I stopped attempting to film myself and concentrate on practicing to fly. I was able to take it up higher than the Animal Sciences buildings, though it wasn't my original intention. I think the wind gave it some additional vertical thrust, and I had a fun time trying to get it back towards me to land safely. Every time it lost altitude over the buildings or trees my heart skipped a beat. I was pleading and bargaining with it in my mind, Please, don't land there, come back over this way. No, not that way! Oh God, please please please come back! If someone were there with me, I'd have gladly said these things out-loud simply for the comedic value.

Through all the fun I had getting back to me, I noticed that, though there was attitude stability, there was almost no yaw stability on the craft. It would start rotating about its vertical axis without cause, which is what made flying it safely back to its takeoff location so difficult. While it was high up in the air, I had a difficult time figuring out where its front was since there weren't any LED's on its underside, plus it was flying in twilight, and its symmetry made the Draganflyer's sillhouette difficult to interpret. As a result, if I knew where its front was for a moment, it would suddenly yaw left or right a few times, and I would become disoriented. I'm sure I will learn to cope with it at some point, but in the mean time, I might consider canabalizing the xUFO for its LED's and mount them on the underside of the Draganflyer. Also, when working on my own Stability Augmentation System (SAS), I'll try to keep yaw in close mind.

TL;DR

• Thermal attitude stability works: while off, craft pitches and rolls with greater magnitude; take video to illustrate this
  
• Buy spare parts: namely rotors and nylon screws, since they seem to take the most of a crash
• Add LED's to bottom of motors/body: take the red LED from the xUFO and mount it on the front motor, and then add a blue LED to the underside of the right motor
• Try to implement yaw stability for SAS: see if it's possible to keep the rotorcraft from yawing on its own or from an external disturbance, while still allowing the pilot to control it; consider a filter like on the F16